Understanding the mechanisms associated with the induction and preservation of immunological memory is needed for the comprehension of how the immune system controls viral infections, how viral infections alter immune system function, and how effective, long-lasting vaccines can best be designed. Our studies with viruses in murine systems have shown that CD8 T cell memory is normally stable, but that this stability is disrupted by other pathogens, whose infections contribute to a loss in memory. Superimposed on this process is the degeneracy of T cell recognition, which allows T cells to often cross-react between unrelated viruses, thereby resulting in an enrichment in cross-reactive T cells in the memory pool. Memory T cells laid down as a consequence of one infection can influence protective immunity and immunopathology associated with a second infection with an unrelated virus, and we have referred to this phenomenon as T cell-dependent heterologous immunity and immunopathology. We have defined new model systems and assembled suitable reagents and molecular techniques to perform an in depth characterization of the evolution of cross-reactive and non-cross-reactive CD8 T cell responses during a series of viral infections and to make unique insights into the life, death, and affinity maturation of CD8 T cell populations in general. This proposal uses several viruses but focuses on lymphocytic choriomenigitis (LCMV) and Pichinde (PV) viruses, Old World and New World arenaviruses whose T cell responses are very well-defined, and on the poxvirus vaccinia (VV), which is used as a vaccine for small pox and as a recombinant vaccine and vector for many other antigens. We propose to explore the evolution of CD8 T cell responses in work supporting the following specific aims: SPECIFIC AIM #1. To determine how the T cell repertoire evolves during sequential infections with acute and persistent viruses encoding cross-reactive epitopes. SPECIFIC AIM #2. To determine which T cell functions occur in the absence of foreign antigen signaling. SPECIFIC AIM #3. To resolve the complex patterns of memory T cell proliferation induced by VV-infection of LCMV-immune mice, as a model for heterologous immunity. SPECIFIC AIM #4. To determine the basis for the deficiency of virus-specific T cells in the homeostatic reconstitution of lymphopenic environments.